By gantisstorm
One-shot swarm commands with Codex CLI planning. Scouts gather context in parallel, planner synthesizes XML architectural instructions for Codex (gpt-5.2), parallel coders execute. No iterative loops - fast parallel execution.
Investigates codebase and returns raw context for planner agents.
Researches external documentation and returns raw context for planner agents.
Implements a single file by parsing full plan for its instructions, then verifies and fixes errors.
Synthesizes context into XML architectural instructions for Codex CLI. Returns full plan for coders.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A pluggable multi-agent planning framework for Claude Code. Swap planning engines (Claude, RepoPrompt, Codex, Gemini) while keeping the same discovery agents and execution patterns.
The Core Insight: All planning engines follow the same meta-pattern: Discovery → Planning → Execution. The framework standardizes discovery (code-scout, doc-scout) and execution (plan-coder) while making the planning layer pluggable. This means you can switch between Claude's speed, RepoPrompt's file intelligence, Codex's reasoning depth, or Gemini's stateless simplicity - all using identical workflows.
┌───────────────────────────────────────────────────────────────────────┐
│ PAIR PLANNING FRAMEWORK │
│ │
│ DISCOVERY (same for all) PLANNING (choose one) │
│ ──────────────────────── ───────────────────── │
│ │
│ ┌─────────────┐ ┌─────────────────────────────┐ │
│ │ code-scout │ │ Claude (pair-*) │ │
│ │ (codebase) │ ├─────────────────────────────┤ │
│ └──────┬──────┘ CONTEXT │ RepoPrompt (repoprompt-*)│ │
│ │ ─────────► ├─────────────────────────────┤ │
│ ┌──────┴──────┐ │ Codex (codex-*) │ │
│ │ doc-scout │ ├─────────────────────────────┤ │
│ │ (external) │ │ Gemini (gemini-*) │ │
│ └─────────────┘ └──────────────┬──────────────┘ │
│ │ │
│ │ PLAN │
│ ▼ │
│ │
│ EXECUTION (same for all) │
│ ──────────────────────── │
│ │
│ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │
│ │ plan-coder │ │ plan-coder │ │ plan-coder │ (parallel) │
│ │ file 1 │ │ file 2 │ │ file 3 │ │
│ └─────────────┘ └─────────────┘ └─────────────┘ │
└───────────────────────────────────────────────────────────────────────┘
The framework provides:
The framework offers two execution patterns that work with any planning engine.
┌────────────────────────────────────────────────────────────────────────┐
│ ITERATIVE DISCOVERY LOOP │
│ │
│ code-scout ───► CHECKPOINT ───► doc-scout ───► CHECKPOINT ───► ... │
│ (background) │ (background) │ │
│ │ User decides │ User decides │
│ ▼ "Add research" ▼ "Complete" │
│ TaskOutput │ TaskOutput │ │
│ │ │ │ │ │
│ └───────────────┴───────────────┴──────────────┘ │
│ │ │
│ context_package │
└───────────────────────────────┼────────────────────────────────────────┘
│
▼
┌────────────────────┐
│ PLANNER │
│ (pluggable) │
└─────────┬──────────┘
│
┌───────────────┼───────────────┐
▼ ▼ ▼
┌────────────┐ ┌────────────┐ ┌────────────┐
│ plan-coder │ │ plan-coder │ │ plan-coder │ (background)
│ file1 │ │ file2 │ │ file3 │
└──────┬─────┘ └──────┬─────┘ └──────┬─────┘
│ │ │
└───────────────┼───────────────┘
▼
TaskOutput
(collect all)
npx claudepluginhub GantisStorm/claude-code-repoprompt-codex-plugins --plugin codex-swarmOne-shot swarm commands with RepoPrompt planning. Scouts gather context in parallel, planner synthesizes XML architectural instructions for RepoPrompt, parallel coders execute. No iterative loops - fast parallel execution.
Multi-agent orchestration for complex coding tasks with Gemini MCP (jamubc/gemini-mcp-tool). Scouts gather raw context, planners synthesize XML architectural instructions for gemini-3-flash-preview. Returns full plan with per-file instructions. Iterative discovery with user checkpoints and parallel execution. Orchestrator manages all context (Gemini is fully one-shot).
Multi-agent orchestration for complex coding tasks with RepoPrompt. Scouts gather raw context, planners synthesize XML architectural instructions with structured sections: task, architecture, selected_context, relationships, implementation_notes, ambiguities. RepoPrompt creates detailed plans with minimal ambiguity. Parallel coders execute with self-verification.
One-shot swarm commands with Codex CLI planning. Scouts gather context in parallel, planner synthesizes XML architectural instructions for Codex (gpt-5.2), parallel coders execute. No iterative loops - fast parallel execution.
Multi-agent orchestration for complex coding tasks. Scouts gather raw context with user checkpoints, planner synthesizes structured narrative (task, architecture, context, relationships, implementation notes, ambiguities) into per-file instructions. Parallel coders execute with self-verification.
Haiku-agent bridge — Claude Code spawns Haiku subagents that run OpenCode via CLI (opencode run). No tmux required. Workers stream colored logs to /tmp/swarm-code-logs/. Saves 70-80% tokens.
Delegate plan execution to Codex CLI via ASP. Part of cc-multi-cli-plugin. Requires the `multi` plugin.
Run Claude Code dynamic-workflow scripts on a local Codex App Server — fan out Codex/GPT agents with the agent/parallel/pipeline DSL, sessionful steerable workers, supervised fleets (--multi), live viewers, and run analytics.
Durable multi-worker coding orchestration with supervised CLIs, metric routing, transactional integration, recovery, and explicit approvals.
Your ChatGPT/Codex subscription models in Claude Code's /model picker. A local gateway routes claude-codex-* models to OpenAI's Codex backend (via claude-code-proxy) and everything else to api.anthropic.com, so you switch between Claude and GPT-5.x mid-session like any other model. No OpenAI API key.
Parallel coding orchestration + Codex image generation — /pumasi for parallel code, /pumasi:image for AI image generation